“Activity Detection” is a basic type of motion detection function (refer to “Motion Detection in General”) incorporated in some Sony surveillance cameras. To detect whether there is motion, Activity Detection compares the luminance level of corresponding pixels in two adjacent picture frames. If the results of this comparison are greater than a preset threshold, it determines there is movement in the monitored area and an alarm is triggered (refer to “Alarm Trigger”). Usually, several detecting areas can be set where attention should be focused, such as office entrances and windows.

The “Activity Detection” function is a convenient feature for surveillance operations. However, it sometimes detects noise as moving content and outputs a false alarm. This is because only two picture frames are used for comparison. For instance, repeated motion patterns such as shaking tree leafs and rippling waves can be incorrectly detected as motion. To reduce such false alarms, more advanced features such as “Motion Detection” (refer to “Motion Detection”) and “Intelligent Motion Detection” (refer to “Intelligent Motion Detection (IMD)”) functions have become available.

Prior to the development of the color video system, experiments in colorimetry proved that most colors visible to the human eye could be composed using the three primary colors, Red (R), Green (G), and Blue (B). This also means that almost any visible color can be separated into a certain combination/amount of the three primary color components. This principle is called “Additive Mixing.” (Fig. A)

The mechanism of reproducing color images on a video monitor is based on this principle and represents a good example for understanding how additive mixing works. Within the video monitor CRT tube, there are three R, G, and B guns. Each gun is used with its associated phosphor, which lights in Red, Green, or Blue when stimulated by the gun’s electron beam (Fig. B). When the CRT tube receives a color video signal, the R, G, and B guns emit electrons (electron beams) towards their phosphors in proportion to the amount of R, G, and B components contained in the signal. This results in the emission of Red, Green, and Blue light, with intensities proportional to their associated electron beams. To the human eye, these lights are perceived as a single ray of light, with the appropriate color reproduced when viewed from a certain distance.

The mechanism of a color video camera uses a reverse function of a video monitor. Light entering the camera lens is first separated into the three primary colors using a prism system (refer to “Prism”) or color filters (refer to “One-Chip Imaging System”). These R, G, and B color lights are then converted into R, G, and B signal voltages at their associated R, G, and B imager sensors (CCD or CMOS) (refer to “Imager (Image Sensor)”). After amplification, the R, G, and B signals are processed into the desired signal format (component, composite, etc) to construct the video output signal.

AGC is a function offered on some professional camcorders that allows images to be captured in dark environments. When enough light cannot be captured with the lens iris fully opened, the camera’s AGC function electrically amplifies the video signal level, increasing and optimizing picture brightness.

AGC is also used in conjunction with the lens iris and CCD shutter to establish an automatic exposure function that offers an expanded exposure range (refer to “TLCS (Total Level Control System)”).

Although convenient, it must be noted that the AGC function also degrades S/N ratio (refer to “S/N (Signal-to- Noise) Ratio”) since noise is amplified together, hence it is not used on high-end cameras and camcorders.

The “Alarm Recording” function is a feature incorporated in surveillance recorders that changes recording quality depending on the monitoring situation. This function can significantly reduce the storage capacity requirements of surveillance recorders.

Under normal conditions, an “Alarm Recording” capable recorder stores images at low quality, or stays on standby, since there is less content to note in such situations. However, when an irregular event occurs (alarm event), the recorder switches to high quality mode and captures images of decisive moments at a high frame rate (refer to “Recording Frame Rate (Recording Refresh Rate)”) and high resolution. By switching recording quality according to the scene’s situation, the “Alarm Recording” function can save a drastic amount of storage capacity.

Recorders with “Alarm Recording” capability can be divided into two types. One type of recorder has alarm input ports to switch the recording mode according to an external alarm trigger (refer to “Alarm Trigger”). After receiving an alarm trigger from a door sensor, PIR sensor*, or camera with motion detection, the recorder switches to high quality recording mode.

The other type of recorder has motion detection capability built-in (refer to “Motion Detection in General”), in addition to the alarm input ports mentioned earlier. This type of recorder detects motion internally and switches to high quality mode without a trigger signal input. The need for motion detection within each camera is therefore eliminated, allowing a less expensive system to be built.

The Alarm recording feature switches recording to high quality mode only during an alarm event. However, high quality images are often required prior to and/or after the alarm – for example, to check how an intruder broke into or left an office. For this reason, pre-alarm recording and post-alarm recording modes are available. These modes are used in conjunction with the alarm recording feature. The pre-alarm recording mode allows images prior to the alarm event to be recorded at the same quality as during the alarm event. This is achieved by buffering images into memory before recording to disk. When an alarm event occurs, the recorder first records the images stored in memory – those before the event – and then records the images during the alarm event. Conversely, postalarm recording allows high quality recording for a specific period after the alarm trigger.

With the “Alarm Recording” function and Pre-/Post-alarm recording modes, users can save significant storage capacity while still retaining the full content and quality required to analyze the scene.

*A PIR sensor is a sensor that detects infrared light. Infrared light has a longer wavelength than visible light and is not perceived by the human eye. As all objects emit infrared light, a PIR sensor reacts to human skin and other objects, even in complete darkness.

As its name indicates, “Alarm Trigger” is a feature used in surveillance systems to initiate actions when irregular incidents occur. For example, switching a recorder to alarm recording mode (refer to “Alarm Recording”), turning on/off a light, sounding an audible alarm, or locking a door.

An alarm trigger can be sent from the video surveillance system to the external security system, or alternatively, from the external security system to the video surveillance system.

The following are examples of how the “Alarm Trigger” function is used in the two cases above.

Sending an alarm trigger from the video surveillance system to the external security system

When a video surveillance system equipped with the motion detection function (refer to “Motion Detection in General”) detects motion in a scene, it outputs a trigger signal from its alarm output ports. This trigger signal is sent to security equipment such as alarm units that control audible alarms, and door relays that control door locks.

Sending an alarm trigger from the external security system to the video surveillance system Door contact is a magnetic switch that detects whether a door is opened or closed. When the door is opened, or the door contact is broken, it sends an alarm signal to the video surveillance system’s alarm input ports. After receiving this alarm signal, the video surveillance system takes actions such as switching the recorder to high quality mode and sending an e-mail to notify security staff of the event.

Other major security equipment compatible with the recorder (alarm input ports) are sensor lights, PIR (Passive Infra Red) sensors*, and glass break detectors.

*A PIR sensor is a sensor that detects infrared light. Infrared light has a longer wavelength than visible light and is not perceived by the human eye. As all objects emit infrared light, a PIR sensor reacts to human skin and other objects, even in complete darkness.

When shooting with a camera, there is a certain range of the image that is captured by the camera and displayed on the picture monitor. This range is called the angle of view, which is measured by the angle between the center axis of the lens to the edges of the image in the horizontal, vertical, and diagonal directions. These are called the horizontal angle of view, vertical angle of view, and diagonal angle of view, respectively. Angle of view becomes narrow when a telephoto lens is used. In contrast, it becomes wider with a wide-angle lens. Consequently, the wider the angle of view, the wider the area of the image captured. A camera’s angle of view also varies depending on the size of the imager (refer to “Imager (Image Sensor)”). This means that 2/3-inch type CCD cameras and 1/2-inch type CCD cameras offer different angles of view for lenses with the same focal lengths.

Angle of view can be calculated from the following equation.

w = 2tan-1 y/2f

w: Angle of view

y: Image size (width of the image in horizontal, vertical, or diagonal direction)

f: Focal length

In general, the word “aperture” refers to an opening, a hole, a crack, or any other type of narrow opening. When used in relation to the mechanism of a lens, it stands for the size of the lens’s opening that determines the amount of light directed to the camera’s imager (refer to “Imager(Image Sensor)”). The diameter of the lens aperture can be controlled by the lens iris (refer to “Iris”), which consists of a combination of several thin diaphragms.

When viewing streaming videos over an IP network, we often experience “drop outs” (breakups in the transmitted images) due to network traffic conditions. In surveillance and monitoring applications, the same situation can be encountered if the network traffic exceeds the network’s bandwidth (refer to “Bandwidth in IP Networks”). To avoid such drop outs, Sony incorporates an Adaptive Rate Control function in its IP network cameras that employs the MPEG-4 compression format.

The ARC function automatically adjusts the compression bit rate and frame rate of the video stream sent from the camera. In order to cope with changing traffic conditions, ARC monitors both the Round Trip Time (RTT) and actual IP packet loss, and optimizes the video bit rate and frame rate accordingly.

Without the ARC function, as the available network bandwidth reduces due to increased traffic, drop outs occur. (Fig. A)

 
 

ATW can be considered an extension of AWB (Auto White Balance) (refer to “AWB (Auto White Balance)”) that adds much more convenience. While AWB is used to set the correct color balance for one particular shooting environment or color temperature (refer to “Color Temperature”), ATW continuously adjusts camera color balance in accordance with any change in color temperature.

For example, imagine shooting a scene that moves from indoors to outdoors. Since the color temperature of the indoor lighting and outdoor sunlight are very different, the white balance must be changed in real time in accordance with the ambient color temperature. With the AWB function this is not practical since the white balance switch would have to be pressed with any change in color temperature, plus the camera would have to frame a grayscale chart or white item.

ATW eliminates such procedures since it is completely automatic – the camera readjusts its white balance each time it detects a change in the ambient color temperature. Simply put, the camera white balance will always follow the change of color temperature. Although convenient, it is also important to note that ATW has its limitations in adjustment accuracy.

“Auto iris” is a convenient function that detects the amount of light entering the lens and automatically opens/closes the iris to maintain appropriate exposure. Auto iris is especially useful in situations where manual iris adjustment can be difficult, such as in ENG apprication & outdoor surveillance.

Auto iris lenses control the iris aperture (refer to “Aperture”) by detecting and analyzing the amplitude of the video signal produced in the camera. An iris control signal is generated according to the amplitude of this video signal, to either open or close the iris for correct exposure.

Unlike the human eye, cameras are not adaptive to different color temperatures (refer to “Color Temperature”) of different light source types or environments. This means that the camera must be adjusted each time a different light source is used, otherwise the color of an object will not look the same when the light source changes. This is achieved by adjusting the camera’s white balance* (refer to “White Balance”) to make a ‘white’ object always appear white. Once the camera is adjusted to reproduce white correctly, all other colors are also reproduced as they should be.

Manual white balance adjustments require technical skills and can be time-consuming. For this reason, professional cameras have an AWB feature that allows white balance to be automatically adjusted simply by the press of a switch. This feature comes in handy when there is no time for manual adjustments or for operators who are not familiar with white balance.

AWB is often mistaken with ATW (refer to “ATW (Auto Tracing White Balance)”), however, the too are quite different. While ATW is ‘completely’ automatic and constantly adjusts the white balance in accordance with the change of the lighting environment, AWB is intended for setting the correct white balance for only one particular environment. Therefore, it must be activated each time there is a small change in the ambient color temperature. This may seem inconvenient; however, AWB achieves a much higher level of accuracy compared to ATW.

AWB is achieved by framing the camera on a white object – typically a piece of white paper/clothe or grayscale chart – so it occupies more than 70% of the display. Then pressing the AWB button on the camera body instantly adjusts the camera white balance to match the lighting environment.

*The correct color conversion filter must also be selected.

Flange-back is an important specification to keep in mind when choosing a lens. Flange-back describes the distance from the camera’s lens-mount plane (ring surface or flange) to the imager’s (refer to “Imager (Image Sensor)”) surface (such as a CCD), as shown in the figure below. In other words, flange-back is the distance that the mounted lens must correctly frame images on the camera’s imager sensor. Therefore, it is necessary to select a lens that matches the flange-back specifications of the given camera.

Flange-back is measured differently depending on whether the camera uses a three-chip or one-chip imaging system (refer to “One-Chip Imaging System” and “Three-Chip Imaging System”). The flange-back of a one-chip camera is simply the distance between the lens mount plane and the imager’s surface. In contrast, the flange-back of a three-chip camera additionally includes the distance that light travels through the prism system (refer to “Prism”) used to separate it into R, G, and B color components. The distance that light travels through this glass material is converted to the equivalent distance if it had traveled through air.

In today’s cameras, flange-back lengths are standardized for each type of lens-mount system. While single CCD cameras use either the C-Mount or CS-Mount system, three-chip cameras use the bayonet mount system. The flange-back of the C-Mount and CS-Mount systems is standardized as 17.526 mm and 12.5 mm, respectively. There are three flange-back standards for the bayonet mount system, 35.74 mm, 38.00 mm, and 48.00 mm.

Similar to flange-back is back focal length, which describes the distance from the very end of the lens (the end of the cylinder that fits into the camera mount opening) to the imager’s surface. The back focal length of the camera is slightly shorter than its flange-back.

Bandwidth in Analog Video Systems

Similar to horizontal resolution (refer to “Horizontal Resolution”), the bandwidth of a video device describes its ability to reproduce image details. However, bandwidth describes this in a very different way. Horizontal resolution indicates a device’s maximum resolving power only, based on the human eye’s visual perception. In contrast, bandwidth describes how much energy can be handled (stored/reproduced) and preserved within a given video frequency range. For this reason, the term bandwidth is used to express the amount of picture detail a storage device can handle, while horizontal resolution is generally used to express the resolving power of a video camera or monitor. A wider bandwidth on a VTR directly translates into images with sharper picture edges and more detail. To understand this, an analysis of the video signal is required.

All video signals (or all signals) are a combination of multiple sine waves with different frequencies. This fact is based on the Fourier transform. Interestingly, a simple video signal with only a black-to-white, white-to-black transition is a combination of the sine waves illustrated in Fig. 1.

The important point to note is that the main rectangular area of this signal is formed with a gradual sine wave (low frequency), while the black-to-white, white-to-black transitions are formed using extremely sharp sine waves (high frequency). Needless to say, these abrupt transitions represent the picture edges of an image. Returning to our discussion, bandwidth describes the frequency range of sine waves that a video device can handle. Fig. 2 shows two video devices with different bandwidths.

As illustrated in this figure, both video devices can handle low-frequency sine waves. This is typical with all video devices. However, as frequency increases, the ability to handle high-frequency sine waves degrades for device (a). In our above analogy using black-to-white and white-to-black transitions, this video device cannot reproduce such transitions sharply, making picture edges appear blurred.

Detailed picture content consists of many abrupt transitions or picture edges. This means that the wider the device’s bandwidth, the sharper the sine waves, and therefore the sharper the picture.

With some inaccuracies, a rough calculation can be made between bandwidth and its equivalent horizontal resolution. Multiplying the bandwidth by 80 gives you the approximate horizontal resolution equivalent. Therefore, with a bandwidth of 6 MHz, the horizontal resolution is approximately 480 TV lines (i.e., 6 MHz x 80). This index is called the factor 80.*

*Applies to SD signals only.

Bandwidth in Digital Video Systems

A digital device’s resolution is usually described by the number of samples or pixels it uses to express an image. For example, the resolution of standard definition video is described as 720 x 480 (NTSC)/720 x 575 (PAL), where the first and second numbers indicate the vertical and horizontal pixel counts, respectively. Unlike analog systems, digital video systems use discrete samples (pixels), each of which preserves the correct amplitude representing that area of the image. This means that the bandwidth of a digital video system is free from the analog principles of high-frequency sine waves falling off, and resolution and bandwidth accurately relate to each other.

The bandwidth of a digital video system is determined by the device’s sampling rate, which defines how many samples are taken per second. Logically, a bandwidth of one-half the system’s sampling rate can be obtained. This fact is based on the Nyquist rate. For example, if the system’s sampling rate is 13.5 MHz, a bandwidth of up to 6.75 MHz can be offered.

Bandwidth in IP Networks

In IT systems such as IP networks and computer systems, bandwidth describes the speed that binary data (1s and 0s) can be transferred or processed. Simply put, the wider the bandwidth, the more information (digital data) can be transferred or processed within a given time.

 

In general, bit rate (sometimes referred to as transfer rate) is used to describe the amount of digital data that a video device or computer device can handle. Bit rate is described by the number of bits processed per second, such as 25 Mb/s, which would mean 25 megabits are processed within one second.

In video or computer technology, bit rate is commonly used to indicate the speed that a device can store data or transfer it over a network.

Since faster bit rates mean more information within a given time, this directly translates into a higher quality of content being stored or transferred.

White balance (refer to “White Balance”) enables cameras to provide correct color by adjusting the camera to reproduce white correctly. However, to ensure accurate color reproduction throughout all video levels, it is important that the red, green, and blue channels are also in correct balance when there is no incoming light. In this state, the camera’s red, green, and blue outputs represent the signal ‘floors’ of the R, G, and B signals, and unless these signal floors are matched, the color balance of other signal levels will not match either. Simply put, without this adjustment, the red, green, and blue color balance cannot be precisely matched even with correct white balance adjustments. This adjustment is called black balance.

Black balance is adjusted by closing the camera lens iris and resetting the red, green, and blue video amplifiers to the same absolute black level – the signal level output when the camera captures no light. With most video cameras, black balance can be achieved by the press of a button thanks to an Auto Black Balance function that automatically closes the lens iris and balances the R, G, and B black levels.